Urban emergency management system and method, and storage medium

Abstract

The invention relates to an emergency management system and method for urban emergencies, computer equipment and a storage medium, and the system comprises a sensing module which is used for collecting urban multi-source heterogeneous data in real time, and carrying out the preprocessing of the multi-source heterogeneous data, so as to generate event situation data; the arrangement module is used for receiving the event situation data, performing event identification based on a preset event processing rule and an emergency knowledge model, and generating an emergency response process; the execution module is used for receiving and analyzing the emergency response process, generating a specific execution instruction and carrying out cooperative scheduling and task allocation on required emergency resources; wherein the execution result data of the execution module is fed back to the arrangement module, and is used for carrying out iterative optimization on the arrangement module. Therefore, the urban emergency management scheme based on dynamic plan generation is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission optimization processing, and in particular to a method, computer equipment and storage medium for urban emergency management. Background Art

[0002] Urban emergency management services are a key mechanism to ensure citizen safety and social stability. They cover the entire process of management from prevention and preparation to response and recovery, and aim to effectively respond to various emergencies including natural disasters (such as earthquakes, floods, and typhoons) and man-made disasters (such as fires, traffic accidents, and terrorist attacks).

[0003] However, traditional urban emergency management, with its static plans based on historical cases, is ill-equipped to handle new, complex incidents. It cannot keep pace with the rapid evolution of emergencies, and responses are often delayed. Cross-departmental command systems are plagued by data silos and collaboration barriers. Summary of the Invention

[0004] Based on this, it is necessary to provide an urban emergency management system, method and storage medium that adopts a dual-channel architecture of "event stream processing + graph reasoning" to realize dynamic generation of plans to address the above problems.

[0005] The first aspect of the present application provides an urban emergency management system, comprising:

[0006] A perception module is used to collect multi-source heterogeneous data in the city in real time and pre-process the multi-source heterogeneous data to generate event situation data;

[0007] An orchestration module is configured to receive the event situation data, identify events based on preset event processing rules and emergency knowledge models, and generate an emergency response process;

[0008] An execution module is used to receive and analyze the emergency response process, generate specific execution instructions, and coordinate the scheduling and task allocation of required emergency resources;

[0009] The execution result data of the execution module is fed back to the orchestration module for iterative optimization of the orchestration module.

[0010] In some embodiments, the perception module includes:

[0011] A multimodal sensor network, deployed in urban surveillance areas, includes at least one of the following sensors: radar water level gauge, smart manhole cover, fiber optic vibration sensor, and infrared thermal imaging camera;

[0012] A hybrid networking unit, including 5G and LoRaWAN heterogeneous network architecture, for data transmission of the multimodal sensor network;

[0013] The data preprocessing pipeline includes a noise filtering unit for filtering out abnormal data and a spatiotemporal alignment unit for fusing data from different sources.

[0014] In some embodiments, the orchestration module includes:

[0015] A composite event processing engine, built on a stream processing framework, is used to detect composite event patterns from the event situation data in real time based on predefined spatiotemporal association rules;

[0016] An emergency knowledge graph, storing entities, relationships, and rules in the emergency field, for performing semantic enhancement and causal reasoning on events detected by the composite event processing engine;

[0017] The process intelligent evolution engine is used to retrieve, cross and mutate from the emergency case library according to the characteristics of the incident to generate an emergency response process that matches the current situation;

[0018] In some embodiments, the process intelligent evolution engine uses an improved genetic algorithm to generate an emergency response process that matches the current situation; the improvement includes:

[0019] The weighted Jaccard coefficient is used to calculate the case similarity to screen individuals to initialize the population;

[0020] Adopting a directed mutation operator guided by the emergency knowledge graph to improve the effectiveness of the mutation operation;

[0021] The improved genetic algorithm uses a multi-dimensional fitness function including time cost, resource utilization and social impact to evaluate the generated emergency response process.

[0022] In some embodiments, the orchestration module further includes a process compliance verification unit, which uses linear temporal logic or TLA+ formalization method to verify the logical correctness and security before the emergency response process is executed.

[0023] In some embodiments, the execution module includes:

[0024] Elastic microservice clusters encapsulate emergency response functions into independently deployable and scalable microservices;

[0025] Intelligent API gateway, used to implement protocol conversion and data format adaptation between the microservice cluster and external heterogeneous business systems;

[0026] The resource collaborative scheduling unit includes a multi-agent game model based on non-cooperative game theory and multiple emergency participants, which is used to solve the optimal allocation plan of emergency resources;

[0027] It includes a multi-agent game model based on non-cooperative game theory, and uses a multi-objective optimization algorithm to coordinate the scheduling and task allocation of required emergency resources;

[0028] Among them, the traffic scheduling of the intelligent API gateway is dynamically optimized using the Q-learning reinforcement learning algorithm; the resource collaborative scheduling unit uses the Shapley value model to quantify resource contribution and the NSGA-II algorithm to solve multi-objective optimization problems.

[0029] In some of the embodiments, a blockchain evidence storage module is also included, which is built based on consortium chain technology and contains a smart contract for storing the key resource scheduling records and task execution status generated by the execution module on the chain in the form of transactions, forming an unalterable emergency log; the smart contract includes a contribution quantification model, which automatically calculates the contribution score of each participating resource based on the emergency log on the chain.

[0030] In some of the embodiments, Hyperledger Fabric technology is used to implement blockchain evidence storage.

[0031] The second aspect of the present application provides a method for urban emergency management, comprising:

[0032] Collecting urban multi-source heterogeneous data in real time and pre-processing the multi-source heterogeneous data to generate event situation data;

[0033] Receive the event situation data, perform event identification based on preset event processing rules and emergency knowledge models, and generate an emergency response process;

[0034] Receive and analyze the emergency response process, generate specific execution instructions, and coordinate the scheduling and task allocation of required emergency resources;

[0035] The module to be orchestrated is iteratively optimized according to the execution result data to form a closed-loop feedback.

[0036] A third aspect of the present application provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, causes the one or more processors to execute the steps of the above-mentioned method for urban emergency management.

[0037] The aforementioned urban emergency management system, methods, and storage media achieve a paradigm shift in city-level emergency response through the deep coupling of business process orchestration technology with emergency management scenarios. Real-time modeling capabilities based on event feature vectors reduce emergency response initiation time from hours to minutes. A decision-making model that integrates fuzzy logic and multi-objective optimization improves cross-departmental resource collaboration efficiency. Protocol conversion middleware supports seamless integration with over 30 urban governance systems, reducing data fusion latency. The knowledge graph's incremental update mechanism enables the dynamic evolution of emergency decision-making knowledge, improving case matching accuracy.

[0038] The composite event detection engine, centered around Complex Event Processing (CEP) technology, builds a multi-dimensional, highly intelligent event recognition and understanding system. Leveraging a pre-defined, refined CEP rule base, the engine covers 12 typical emergency scenarios and enables spatiotemporal correlation analysis across data sources. The composite event detection engine integrates multi-source heterogeneous data, integrating sensor data, social media information, and other multi-dimensional inputs; dynamic event pattern recognition, accurately capturing the characteristics of emergencies based on spatiotemporal correlation rules; and intelligent risk assessment, automatically triggering appropriate emergency response mechanisms based on event attributes.

[0039] The semantic enhancement mechanism, based on knowledge graph reasoning technology, builds a dynamically evolving emergency semantic understanding system. The semantic enhancement module utilizes a knowledge graph based on the TransH relational reasoning model, a real-time streaming data update mechanism, and multi-dimensional information integration (disaster characteristics, geographic data, and historical cases) to dynamically optimize the emergency decision-making knowledge base.

[0040] Among them, the process semantic extension is based on the standard BPMN 2.0 model. By embedding special emergency semantic tags, it can accurately express the specific needs of emergency management, dynamically embed urgency and resource constraint information, provide legal and regulatory basis, and support condition-based dynamic adjustment of processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a block diagram of the internal structure of a computer device in one embodiment;

[0042] Figure 2 A block diagram of a method for managing urban emergencies in one embodiment;

[0043] Figure 3 A knowledge graph structure of a method for urban emergency management in one embodiment;

[0044] Figure 4 A flowchart of a method for managing urban emergencies in one embodiment;

[0045] Figure 5It is a structural diagram of a computer system suitable for implementing the electronic device of the embodiment of the present application. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the present invention more apparent, exemplary embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments of the present invention, and it should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described in the present invention, all other embodiments obtained by those skilled in the art without creative work should fall within the scope of protection of the present invention.

[0047] Among related technologies, IDC predicts that global smart city investment will reach $327 billion in 2025, with emergency management accounting for 18%. The demand for real-time monitoring of urban operational indicators (such as traffic flow, pipe network pressure, and crowd heat) is driving the emergence of dynamic response mechanisms (cited in the "Smart City White Paper 2023").

[0048] However, traditional contingency management suffers from numerous pain points: Response lag: Static contingency plans based on historical cases struggle to cope with new, complex events (such as the COVID-19 pandemic compounded by extreme weather) and fail to keep pace with the rapid evolution of emergencies. Collaboration barriers: Cross-departmental command systems suffer from data silos. For example, data sharing delays of up to 45 minutes between civil affairs, medical, and transportation systems lead to resource scheduling duplication rates as high as 32%.

[0049] To address these issues, business process orchestration can be employed. Its benefits include: Dynamic Adaptation: BPMN 2.0 extensions enable real-time mapping of events, processes, and resources, demonstrating a 60% increase in process reorganization efficiency during Shenzhen's epidemic prevention and control efforts. Digital Twin Empowerment: Integration with the City Information Model (CIM) platform supports simulation and pre-verification of emergency scenarios, reducing implementation risks.

[0050] In one embodiment, the implementation environment of a method for urban emergency management includes a computer device and a server. The computer device is a configuration device, such as a computer, which has a carbon neutrality control tool installed on it. When control is required, a request for urban emergency management can be sent to the server. The server receives the carbon neutrality control request and dynamically manages the equipment in the industrial park. It should be noted that the computer device can be a smartphone, tablet computer, laptop computer, desktop computer, etc., but is not limited to these.

[0051] To facilitate understanding, the following first introduces the relevant terms involved in the embodiments of this application.

[0052] Genetic Algorithm (GA) is an optimization algorithm based on natural selection and genetic mechanisms, often configured to solve complex optimization problems.

[0053] Transfer learning (TL) is a machine learning method that uses a model developed for task A as a starting point and reuses it in the process of developing a model for task B. This means applying the knowledge gained from one task (the source task) to a different but similar task (the target task). For example, the knowledge gained when learning to classify encyclopedia text can be applied to solve legal text classification problems, or the knowledge gained when learning to classify cars can be used to identify birds in the sky. Therefore, when data labels for some tasks are difficult to obtain, transfer learning can be used from other tasks similar to the task for which labels are easily available. Alternatively, to avoid the complexity and time-consuming process of building a model from scratch, transfer learning can be used to accelerate learning efficiency.

[0054] Figure 2 FIG. 1 is a structural diagram of a system for urban emergency management in one embodiment. Figure 2 As shown, the urban emergency management system of this embodiment includes:

[0055] The perception module is used to collect urban multi-source heterogeneous data in real time and pre-process the multi-source heterogeneous data to generate event situation data;

[0056] The orchestration module is used to receive event situation data, identify events based on preset event processing rules and emergency knowledge models, and generate emergency response processes;

[0057] The execution module is used to receive and analyze the emergency response process, generate specific execution instructions, and coordinate the scheduling and task allocation of required emergency resources;

[0058] The execution result data of the execution module is fed back to the orchestration module for iterative optimization of the corresponding orchestration module, forming a closed-loop feedback.

[0059] In some embodiments, the perception module includes:

[0060] A multimodal sensor network, deployed in urban surveillance areas, includes at least one of the following sensors: radar water level gauge, smart manhole cover, fiber optic vibration sensor, and infrared thermal imaging camera;

[0061] Hybrid networking unit, including 5G and LoRaWAN heterogeneous network architecture, for data transmission in multimodal sensor networks;

[0062] The data preprocessing pipeline includes a noise filtering unit for filtering out abnormal data and a spatiotemporal alignment unit for fusing data from different sources; the noise filtering unit uses the density-based DBSCAN clustering algorithm, and the spatiotemporal alignment unit adopts the Kalman filter model.

[0063] The coordinated deployment of multimodal sensing devices is shown in the table below. Based on the urban geographic space grid division (1km×1km), the device density and type are dynamically adjusted. The coordinated deployment of multimodal sensing devices is shown in the table below:

[0064]

[0065] The 5G+LoRaWAN heterogeneous network architecture addresses coverage blind spots. 5G base stations handle high-bandwidth data transmission, such as video streams and laser point clouds, with end-to-end latency ≤ 50ms. LoRa relays build mesh networks in areas like tunnels and underground pipeline corridors, with power consumption less than 1W per node.

[0066] The data preprocessing pipeline includes a noise filtering module based on the improved DBSCAN clustering algorithm and a spatiotemporal alignment unit based on Kalman filtering. The noise filtering unit eliminates outliers based on the improved DBSCAN clustering algorithm. The formula is as follows:

[0067] Cluster(D,e,MinPts)={CIVp∈C,|Ne(p)|≥MinPts}

[0068] Among them, e=3, MinPts=5, and the density parameters are dynamically adjusted to cope with sudden data.

[0069] Spatiotemporal alignment unit: defines a unified spatiotemporal reference system (WGS84 coordinate system + UTC timestamp) and uses Kalman filtering to achieve multi-source data fusion.

[0070] The system comprises a three-tiered architecture: a perception layer deploying a multimodal sensor network (a global IoT perception network), an orchestration layer, and an execution layer. 5G and LoRaWAN heterogeneous networking technologies are used to collect real-time city operational data, with noise filtering performed using an improved DBSCAN algorithm. The orchestration layer includes a dynamic process engine extending BPMN 2.0 (Business Process Model and Notation 2.0), triggering dynamic process generation through event feature vectors (E = (T, S, R, C, P, I)), supporting real-time plug-and-play and compliance verification of process fragments. The execution module, based on elastic microservice clusters and intelligent API gateways, implements heterogeneous system protocol conversion and resource scheduling optimization, employing an improved genetic algorithm and a multi-objective game model for resource scheduling. The knowledge graph module integrates a streaming data update mechanism with the TransH relational reasoning model to dynamically optimize the emergency decision-making knowledge base. The blockchain evidence storage module records the resource allocation process through Hyperledger Fabric and employs a layered consensus architecture to ensure data security.

[0071] In some embodiments, the orchestration module includes:

[0072] The composite event processing engine, built on a stream processing framework, is used to detect composite event patterns from event situation data in real time based on predefined spatiotemporal association rules;

[0073] Emergency knowledge graph, which stores entities, relationships, and rules in the emergency field and is used to perform semantic enhancement and causal reasoning on events detected by the composite event processing engine;

[0074] The process intelligent evolution engine is used to retrieve, cross-reference, and mutate the emergency case library based on the characteristics of the event to generate an emergency response process that matches the current situation;

[0075] The process intelligent evolution engine uses an improved genetic algorithm. The improvements include:

[0076] The weighted Jaccard coefficient is used to calculate the case similarity to screen individuals to initialize the population;

[0077] A directed mutation operator guided by the emergent knowledge graph is adopted to improve the effectiveness of the mutation operation.

[0078] Among them, the adaptive process orchestration engine, as the core decision-making component of the emergency management system, realizes the dynamic generation and optimization of emergency processes through the collaboration of three-layer modules.

[0079] The composite event detection engine, with complex event processing (CEP) technology at its core, accurately identifies emergency event patterns through preset rules, providing support for emergency decision-making. The CEP rule base it has constructed covers 12 typical emergency event patterns, including scenarios such as explosions in hazardous gas plants and urban flooding caused by heavy rain. The engine can conduct multi-source data fusion analysis based on spatiotemporal correlations, such as combining data collected by sensors with social media information, thereby triggering early warnings and achieving accurate identification and in-depth understanding of emergencies. The semantic enhancement module uses knowledge graph reasoning technology to further deepen the semantic understanding of events. Among them, the knowledge graph module adopts a streaming data update mechanism and the TransH relational reasoning model to integrate multi-dimensional information such as disaster characteristics, geographic data, and historical cases in real time, and dynamically optimize the emergency decision-making knowledge base. This process realizes the upgrade from single event detection to multi-dimensional association reasoning, providing more accurate semantic support for emergency process adjustments.

[0080] The synergy between the two enables the system to automatically infer the measures that should be taken based on the semantic relationship of events. For example, when an event of type "heavy rain" with a risk level greater than 8 is detected, a trigger relationship will be automatically created to associate the action of "initiating a level I flood prevention response", greatly improving the efficiency and accuracy of emergency response.

[0081] The emergency metamodel, expanded upon standard BPMN 2.0, enhances its applicability in emergency management by embedding emergency semantic tags, including criticality (urgency), resourceConstraint (resource constraint), and legalBasis (legal basis). This expanded emergency metamodel accurately expresses the unique requirements of emergency management: the criticality tag clearly demonstrates the urgency of an event, the resourceConstraint tag clarifies resource requirements, and the legalBasis tag provides the corresponding legal and regulatory basis. Furthermore, the metamodel supports dynamic process adjustments based on emergency conditions.

[0082] This extended design enables the emergency metamodel to accurately configure and support the precise description of emergency response processes based on the BPMN2.0 standard, allowing the emergency management system to integrate the special information and processing logic required for emergency scenarios into the standardized process framework, thereby achieving more flexible and targeted emergency response process management.

[0083] A distributed process repository uses GitOps to version control and manage emergency response plans. These branches include Main, Feature, and Hotfix branches. The Main branch contains verified, stable plans (ISO22301 certified). The Feature branch contains experimental processes for new incidents. The Hotfix branch fixes configuration errors in running processes.

[0084] The emergency response management system, built using the GitOps philosophy, applies version control methods from modern software development to the field of emergency management. Specifically, GitOps is an operational model that uses Git as a single source of truth, managing infrastructure and application configurations through a version control system. In the emergency response management system, this means that all emergency response plans, processes, and configurations are stored as code in a Git repository, enabling traceability, rollback, and collaborative development.

[0085] Three branch types and their functions:

[0086] The Main branch is the branch that stores verified and stable emergency plans. These plans have been rigorously tested and verified and can be directly applied to actual emergency responses.

[0087] Feature branches are used to develop experimental processes for new or unconventional events. In feature branches, you can safely innovate and experiment with contingency plans without affecting the verified processes in the main branch. Once these experimental processes are verified, they can be merged into the main branch to become standard plans.

[0088] Hotfix branches are specifically used to perform emergency repairs on running processes, allowing configuration errors or vulnerabilities in the plan to be corrected without interrupting system operation.

[0089] This "online repair" mechanism ensures that even if problems are discovered during the emergency response process, adjustments can be made quickly without affecting the overall operation. After the repair is completed, it can be immediately merged back to the Main branch to ensure that the plan in the main branch is always up to date and correct.

[0090] Through this distributed process warehouse management, the emergency management system can achieve efficient and reliable plan development and maintenance processes like modern software development, enhancing the emergency response capabilities to various emergencies.

[0091] In some embodiments, the improved genetic algorithm uses a multi-dimensional fitness function including time cost, resource utilization and social impact to evaluate the generated emergency response process.

[0092] The genetic algorithm is an optimization algorithm that simulates the natural evolution process. Here, it is specifically improved for the generation of emergency processes. The chromosome encoding scheme represents the emergency process as a "chromosome" that the genetic algorithm can process:

[0093] Header: Contains a 4-bit event type code (which can represent 16 different event types) and a 2-bit priority code (which can represent 4 priority levels). This information determines the overall positioning and importance of the process.

[0094] The Body is a variable-length sequence of genes, each of which represents a BPMN node (task, gateway, or event) in the process. This design can represent arbitrarily complex process structures.

[0095] Footer part: Use a 16-bit mask to indicate resource constraints, such as the minimum required quantity or availability conditions of each resource.

[0096] This encoding scheme contains both structural information of the process and metadata such as event types and resource constraints, enabling the genetic algorithm to optimize the process while considering actual constraints.

[0097] Crossover operator optimization, a "semantics-preserving crossover" algorithm, solves the problem that the crossover operation of traditional genetic algorithms may cause process structure destruction:

[0098] First, find the matching gateways in the two parent processes (such as the start and end pairs of parallel gateways), and exchange process fragments between these matching points to ensure that the exchanged fragments are structurally meaningful.

[0099] Verification is performed after the crossover to ensure that the generated child process has complete start and end events and logic. This method can ensure that the crossover operation will not destroy the basic structure and semantics of the process and avoid generating invalid emergency processes.

[0100] The urban emergency management method uses an improved weighted Jaccard method to calculate case similarity. The traditional Jaccard similarity coefficient has limitations because it only considers the intersection and union ratios of feature sets, ignoring the varying importance of different features in emergency management. For example, the "size of the affected population" is clearly more important than the "number of affected administrative regions." The improved method introduces a weighted Jaccard coefficient, with weights determined using the Analytic Hierarchy Process (AHP). See the table below:

[0101] Feature Item Weight Decision matrix consistency test CR = 0.032 Event Type 0.35 λ_max=5.21 Scope of impact 0.28 CI = 0.052 Resource gap rate 0.18 RI=1.12 Time urgency 0.12 CR=CI / RI=0.046<0.1 Political sensitivity 0.07

[0102] Improved genetic algorithm designs include: Optimized population initialization: The random initialization method of traditional genetic algorithms is inefficient. This system is improved to use similarity screening based on the case library, prioritizing historical solutions similar to the current situation as the initial population to accelerate algorithm convergence.

[0103] Directed mutation operator: Combined with the emergency management knowledge graph to guide the mutation direction, such as:

[0104] When the "Insufficient Medical Resources" problem is detected, the probability of inserting the "Activate Fangcang Hospital" node is increased by 40%;

[0105] When a "traffic disruption" problem is detected, the probability of replacing the "drone delivery" branch increases by 30%; this intelligent variation greatly improves the efficiency of the algorithm.

[0106] In some embodiments, the orchestration module further includes a process compliance verification unit, which uses linear temporal logic or TLA+ formal methods to verify the logical correctness and security of the emergency response process before execution.

[0107] In some embodiments, the execution module includes:

[0108] Elastic microservice clusters encapsulate emergency response functions into a series of independently deployable and scalable microservices;

[0109] Intelligent API gateway, used to implement protocol conversion and data format adaptation between microservice clusters and external heterogeneous business systems;

[0110] The resource collaborative scheduling unit builds a multi-agent game model involving multiple emergency participants to solve the optimal allocation plan for emergency resources;

[0111] As a key component of the execution layer, the intelligent API gateway plays an important role in heterogeneous system protocol conversion and resource scheduling optimization. Its configuration and functional design are fully adapted to the needs of emergency management scenarios. The gateway supports a variety of heterogeneous system interface conversions. On the one hand, it can achieve efficient conversion of 30+ data formats (such as XML, JSON, Protobuf). On the other hand, it can complete transparent bridging of communication protocols such as HTTP / HTTPS, CoAP, MQTT, and effectively break down data interaction barriers between different systems. In terms of routing optimization, the intelligent API gateway uses the Q-learning reinforcement learning algorithm to dynamically optimize the routing path, and adjusts the data transmission path in real time based on six-dimensional indicators such as current load, latency, and error rate. It can significantly reduce transmission delays, improve data transmission efficiency, ensure the smooth flow of emergency instructions and data in complex systems, and provide stable communication support for emergency resource scheduling and task execution. The intelligent API gateway also adopts a service classification strategy to further improve the accuracy and efficiency of service calls during the emergency response process. At the same time, the collaborative resource scheduling unit uses the Shapley value model to quantify resource contributions and solves multi-objective optimization problems through the NSGA-II algorithm, jointly building an efficient resource scheduling and communication support system with the intelligent API gateway. The intelligent API gateway also adopts a service classification strategy, as shown in the table below:

[0112]

[0113]

[0114] In some embodiments, a blockchain evidence storage module is also included, which is built based on consortium chain technology and contains a smart contract for storing key resource scheduling records and task execution status generated by the execution module on the chain in the form of transactions, forming an unalterable emergency log; the smart contract includes a contribution quantification model, which automatically calculates the contribution score of each participating resource based on the emergency log on the chain.

[0115] In some embodiments, Hyperledger Fabric technology is used to implement blockchain evidence storage, and emergency contribution is automatically recorded and quantified by deploying smart contracts.

[0116] In some embodiments, the orchestration module further includes:

[0117] The process fragment dynamic plug-in unit is configured to dynamically adjust the process composition according to the characteristics of the emergency event; the event characteristics are used as the conditions for triggering rule matching, and the rules in the rule library are matched according to the event characteristics, and the process is adjusted according to the definition of the rules;

[0118] Mutation rule library, which stores preset process mutation conditions, mutation operations, and their effective time;

[0119] Hot-swap unit for versions, based on Java Instrumentation, enables zero-downtime updates through differential updates.

[0120] The process compliance verification unit uses the TLA+ formal method to check process attributes and automatically associate relevant legal clauses.

[0121] The process variation mechanism includes the dynamic insertion and removal of process fragments and the variation rule library. The dynamic insertion and removal of process fragments means that according to the changing situation, the system can dynamically insert, remove or replace certain links in the process, so that the emergency process can adapt to the changing situation. The variation rule library defines a set of rules for trigger conditions, variation operations and effective time. For example, this rule library defines how the emergency process should be changed under specific conditions. For example, when the occupancy rate of hospital beds exceeds 90%, the system will automatically insert the process of starting the cabin hospital within 15 minutes; when the main roads are blocked, the system will quickly switch to the drone delivery solution; when there is a shortage of industry personnel, AI assistance will be immediately enabled. The variation rule library is shown in the table below:

[0122] Trigger Conditions mutation operation Effective Time Hospital bed occupancy rate >90% Insert the Fangcang Hospital startup process 15 minutes Number of main road closures ≥ 3 Replaced with drone material delivery solution 5 minutes Emergency expert online rate <30% Enable AI-assisted decision-making module Effective immediately

[0123] The system dynamically updates running program code without shutting down. For emergency response systems, this means that processing logic can be updated in real time during the emergency response process without interrupting the currently executing emergency process. The correctness of the process is verified using the TLA+ (Temporal Logic Action) language. It stipulates an invariant: for all tasks in the "Executing" state, there must be an available resource allocated to the task. This ensures that the process design is logical and avoids unexecutable process steps.

[0124] By binding legal terms, the system automatically links relevant laws and regulations (such as the Emergency Response Law) to ensure that the emergency response process complies with legal requirements and improves the legitimacy and authority of decision-making.

[0125] Exemplarily, the resource scheduling optimization method includes:

[0126] A two-stage matching algorithm that quickly screens candidate resources based on an R-tree index and solves for optimal allocation using the Hungarian algorithm.

[0127] Multi-objective game model, using Shapley value calculation and NSGA-II algorithm to generate Pareto front solution set;

[0128] Blockchain smart contracts automatically execute resource allocation records and contribution scores.

[0129] Protocol conversion middleware includes:

[0130] Precompiled template technology supports efficient conversion of more than 30 data formats including XML, JSON, Protobuf, etc.

[0131] Q-learning routing optimization model, dynamically adjusting data transmission paths to reduce latency;

[0132] Service grading strategy allocates computing resources according to response time requirements (L0: ≤ 100ms, L1: ≤ 1s, L2: ≤ 10s).

[0133] Exemplarily, the incremental update method of the knowledge graph module includes:

[0134] Entity disambiguation algorithm, which calculates semantic similarity based on the improved Levenshtein distance;

[0135] Four-stage update process, including streaming data collection, entity disambiguation, TransH relationship reasoning, and semantic difference version control;

[0136] Knowledge freshness index, dynamically eliminates outdated knowledge items through an exponential decay factor.

[0137] In some embodiments, an improved weighted Jaccard similarity calculation model is used to evaluate the feature weight differences of cases by introducing feature importance vectors; wherein, a hierarchical analysis method is used to determine the feature item weights and perform a consistency check.

[0138] Elastic microservice clusters allocate resource quotas and manage the number of replicas according to preset service levels;

[0139] Service governance framework, including service tiering strategy and Hystrix-based circuit breaker configuration;

[0140] Fuse configuration: Fault isolation is implemented based on Hystrix, threshold setting: requestVolumeThreshold: 20 (request volume threshold), this parameter sets the minimum number of requests that trigger the fuse judgment, which means: there must be at least 20 requests within the statistical time window (the default is 10 seconds) before the fuse considers whether it needs to be opened. For example: if only 15 requests come within 10 seconds, even if all fail, the fuse will not open because the sample size is insufficient, errorThresholdPercentage: 50% (error percentage threshold), this parameter sets the error rate threshold for activating the fuse. The meaning is: when the proportion of failed requests to the total requests reaches or exceeds 50%, the circuit breaker will open. For example: if there are 30 requests within 10 seconds, of which 16 fail (a failure rate of about 53%), the circuit breaker will open. SleepWindow: 5000 (sleep time window) sets the recovery attempt waiting time after the circuit breaker is opened, the unit is milliseconds, which means: after the circuit breaker is opened, all requests will be rejected within 5000 milliseconds (5 seconds). After 5 seconds, the circuit breaker will enter a "half-open" state, allowing one request to pass to test whether the service has recovered. The main purpose of this mechanism is to protect the system from cascading failures and ensure that the failure of a service does not cause the entire system to crash. It is an important means to achieve high availability in a microservice architecture. By automatically cutting off the request flow of the failed service, the system can continue to operate with some functional degradation instead of being completely paralyzed.

[0141] The incident priority assessment model uses a multidimensional assessment matrix to determine the urgency and priority of an incident. It also uses a cellular automation-based situational analysis algorithm to simulate the possible spread and impact of an incident. This enables the system to predict incident development trends and prepare for resource allocation in advance.

[0142] The resource binding strategy uses a multi-objective optimization model to balance resource allocation under various constraints, such as time efficiency, resource utilization, and coverage. The system uses Deb's rule to convert various constraints into penalty terms and comprehensively optimize the resource allocation plan.

[0143] Resource allocation is recorded using the Hyperledger Fabric blockchain platform. When the system allocates resources, this smart contract is invoked, updating the resource status to "allocated" and recording this action on the blockchain. This ensures transparency and immutability of the resource allocation process, facilitating subsequent audits and accountability.

[0144] Resource Collaboration Multi-Objective Game Model: Multi-Agent Game Scenario Setting:

[0145] Emergency response usually involves collaboration among multiple departments, and the system has established a three-party multi-objective game model:

[0146] Participants: government departments (G), medical institutions (M), logistics companies (L);

[0147] Resource pool: {Ambulance A1-A2, Doctor B1-B3, Material Warehouse C1};

[0148] Utility function definition:

[0149] Ug=0.5Q rescue + 0.3Q order - 0.2C cost

[0150] UM = 0.7Q treatment - 0.3C manpower

[0151] UL=0.6Q for delivery-0.4C for transportation

[0152] UG=45,Um=32,UL=28

[0153] An example of Shapley value calculation, considering the marginal contribution calculation of ambulance A1, is shown in the following table:

[0154]

[0155] Nash equilibrium point: When G chooses the middle input, M distributes B1+B2, and L chooses the shortest path, the three parties’ utilities reach a balance:

[0156] UG=45,Um=32,UL=28

[0157] Pareto front: The non-dominated solution set is obtained through the NSGA-II algorithm, and the compromise solution is selected as the final solution.

[0158] The system has designed a blockchain-based contribution notarization model to record the contributions of resource providers. This design ensures the transparency and non-tamperability of resource contribution records.

[0159] For example, contribution records are uploaded to the chain through smart contracts. The system also designs a dynamic scoring mechanism to calculate the final score based on service time and user evaluation, and automatically triggers the audit process when the score is too low to ensure service quality.

[0160] The system adopts a main chain (Hyperledger Fabric) and side chain (Ethereum) architecture, and realizes cross-chain verification through relay bridge. At the same time, it optimizes the PBFT consensus algorithm and reduces the communication complexity from O(N 2 ) is reduced to O(N), greatly improving the system processing capacity.

[0161] Technical verification data shows that the system maintains an average latency of 520ms and a throughput of 1923 TPS under 1000 concurrent requests. Security verification demonstrates that even in a scenario where 10% of nodes are malicious, the system maintains 100% data consistency and double-spending resistance. The blockchain evidence storage mechanism ensures transparency and traceability of the emergency response process.

[0162] It is understandable that multi-source data fusion: integrating multi-source data such as sensor data, social media information, medical records, etc.; spatiotemporal correlation analysis: considering the time and geographical location of events at the same time; knowledge-enhanced reasoning: using domain knowledge to make up for the limitations of pure data-driven; predictive identification: not only identifying events that have occurred, but also predicting potential risks; in urban emergency management, this set of algorithms can help decision makers identify risks earlier and more accurately, provide technical support for preventive emergency management, and ultimately reduce the losses caused by emergencies.

[0163] Process compliance verification algorithm and formal verification model generated by emergency process

[0164] Model checking toolchain: Integrates the NuSMV verifier and uses formal methods to verify that the process conforms to the expected behavioral specifications.

[0165] Runtime monitoring mechanism, dynamic constraint checking mechanism.

[0166] It is understandable that the urban emergency management method integrates event detection, process optimization, resource scheduling, and blockchain evidence storage into a complete system that works together.

[0167] Algorithm interaction protocol: Closed-loop data flow design:

[0168] A[CEP]-->|Trigger signal|B[Process genetic optimization]

[0169] B-->|Candidate Solution|C[Resource Game Scheduling]

[0170] C-->|Execution Log|D[Blockchain Evidence]

[0171] D-->|Feedback Data|A

[0172] Event detection (CEP): As the starting point of the process, it is responsible for identifying sudden events from multi-source data;

[0173] Process genetic optimization: After receiving the event trigger signal, the improved genetic algorithm is used to generate the emergency process plan that best suits the current situation;

[0174] Resource game scheduling: Receive candidate process plans and use game theory models to allocate optimal resources to the processes;

[0175] Blockchain evidence storage: The execution process and results are permanently recorded through blockchain technology to ensure that they cannot be tampered with;

[0176] Closed-loop feedback: Historical data stored on the blockchain is fed back to the event detection module to improve future event recognition capabilities;

[0177] This closed-loop design enables the system to self-learn and continuously optimize, and each emergency response will become an empirical data performance tuning strategy to improve future performance.

[0178] The computing resource allocation table shows the system's resource optimization configuration based on the characteristics of each module. See the table below:

[0179] Algorithm Module Number of CPU cores Memory quota GPU acceleration Composite event processing engine 8 32GB no Genetic Algorithm 4 16 GB Yes (CUDA) Blockchain consensus 2 8GB no

[0180] Resource allocation logic includes:

[0181] Composite event processing engine: allocates the most CPU and memory resources because it needs to process high-throughput real-time data streams and has extremely high requirements for real-time response.

[0182] Genetic Algorithm: GPU acceleration is allocated because the population evolution calculation of the genetic algorithm is highly parallel and suitable for execution on the GPU.

[0183] Blockchain consensus: allocates fewer resources because it mainly ensures the immutability of data and has relatively low computing performance requirements.

[0184] This differentiated resource allocation ensures the best overall system performance with limited resources.

[0185] Real-time guarantee mechanisms include:

[0186] The time constraint model assigns deadlines to different types of tasks, ensuring that the system can handle tasks in an orderly manner in emergency situations. Preemptive scheduling: The system continuously monitors the task queue and always selects the highest-priority task. If a high-priority task's deadline is approaching (the current time + the task's estimated execution time has exceeded the deadline), the currently executing task is immediately preempted. This ensures that the most urgent tasks are processed first, preventing critical tasks from missing their processing window due to resources being occupied by lower-priority tasks.

[0187] This embodiment achieves a paradigm shift in city-level emergency response through the deep coupling of business process orchestration technology with emergency management scenarios. Real-time modeling capabilities based on event feature vectors reduce emergency response initiation time from hours to minutes. A decision-making model that integrates fuzzy logic and multi-objective optimization improves cross-departmental resource collaboration efficiency. Protocol conversion middleware supports seamless integration of over 30 urban governance systems, reducing data fusion latency. The knowledge graph's incremental update mechanism enables the dynamic evolution of emergency decision-making knowledge, improving case matching accuracy.

[0188] Figure 4 FIG. 1 is a flow chart of a system for urban emergency management in one embodiment. Figure 4 As shown, the urban emergency management method of this embodiment includes:

[0189] S401, real-time collection of multi-source heterogeneous data in the city, pre-processing of the multi-source heterogeneous data to generate event situation data;

[0190] S402: Receive event situation data, identify events based on preset event processing rules and emergency knowledge models, and generate emergency response processes;

[0191] S403: Receive and analyze the emergency response process, generate specific execution instructions, and coordinate the scheduling and task allocation of required emergency resources;

[0192] Among them, the corresponding orchestration module is iteratively optimized according to the execution result data to form a closed-loop feedback.

[0193] In some embodiments, the process intelligent evolution algorithm is an improved genetic algorithm, and its fitness function comprehensively evaluates the time efficiency, resource consumption and expected social impact of the process solution.

[0194] In some embodiments, when the improved genetic algorithm is initialized, the weighted Jaccard similarity between the current event and the historical cases is calculated, and the historical process solutions with high similarity are preferentially selected as the initial population.

[0195] In some embodiments, when the improved genetic algorithm performs a mutation operation, its mutation direction and probability are guided by the domain knowledge and rules stored in the emergency knowledge graph.

[0196] In some embodiments, the multi-objective optimization model is constructed as a multi-agent game model, and the scheduling solution is determined by solving the Nash equilibrium point of the game model.

[0197] In some embodiments, Hyperledger Fabric technology is used to implement blockchain evidence storage, and emergency contribution is automatically recorded and quantified by deploying smart contracts.

[0198] Reference below Figure 5 , which shows an electronic device (eg Figure 1 A schematic diagram of the structure of a computer system of the server shown in FIG. Figure 5 The electronic device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0199] like Figure 5 As shown, the computer system includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 502 or the program loaded from the storage part 508 into the random access memory (RAM) 503. Various programs and data required for system operation are also stored in the RAM 503. The CPU 501, ROM 502, and RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0200] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, and the like; an output section 507 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 508 including a hard disk; and a communication section 509 including a network interface card such as a LAN card or a modem. The communication section 509 performs communication processing via a network such as the Internet. A drive 620 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 620 as needed, so that a computer program read therefrom can be installed into the storage section 508 as needed.

[0201] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code configured to execute the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 509, and / or installed from the removable medium 511. When the computer program is executed by the central processing unit (CPU) 501, the above-mentioned functions defined in the method of the present application are executed. It should be noted that the computer-readable medium of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such a propagated data signal may take a variety of forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transfer a program configured for use by or in conjunction with an instruction execution system, apparatus, or device. The program code embodied on a computer-readable medium may be transmitted using any suitable medium, including, but not limited to, wireless, wire, optical cable, RF, etc., or any suitable combination thereof.

[0202] In another embodiment, the present invention provides a computer-readable medium having a computer program stored thereon, which, when executed, performs the method of the above-described embodiment. Any tangible, non-transitory computer-readable medium may be used, including magnetic storage devices (hard disks, floppy disks, etc.), optical storage devices (CD-ROMs, DVDs, Blu-ray discs, etc.), flash memory, and / or the like. These computer program instructions may be loaded onto a general-purpose computer, a special-purpose computer, or other programmable data processing device to form a machine, such that the instructions executed on the computer or other programmable data processing device can generate a device that implements a specified function. These computer program instructions may also be stored in a computer-readable memory, which can instruct the computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer-readable memory can form an article of manufacture, including an implementation device that implements the specified function. The computer program instructions may also be loaded onto a computer or other programmable data processing device, thereby executing a series of operational steps on the computer or other programmable device to produce a computer-implemented process, such that the instructions executed on the computer or other programmable device can provide steps configured to implement the specified function.

[0203] Although example embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above example embodiments are merely illustrative and are not intended to limit the scope of the present invention. Various changes and modifications may be made therein by those skilled in the art without departing from the scope and spirit of the present invention. All such changes and modifications are intended to be included within the scope of the present invention as claimed in the appended claims.

[0204] In the description provided herein, numerous specific details are described. However, it is understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0205] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the description of exemplary embodiments of the present invention, the various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this approach to the present invention should not be interpreted as reflecting the intention that the claimed invention requires more features than those explicitly recited in each claim. More precisely, as reflected in the corresponding claims, the inventive point is that the corresponding technical problem can be solved with fewer features than all the features of a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into the detailed description, with each claim itself serving as a separate embodiment of the present invention.

[0206] It will be understood by those skilled in the art that, except where mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or apparatus disclosed herein may be combined in any combination. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature providing the same, equivalent, or similar purpose.

[0207] Furthermore, those skilled in the art will appreciate that although some embodiments herein include certain features included in other embodiments but not other features, combinations of features from different embodiments are intended to be within the scope of the present invention and to form different embodiments. For example, in the claims, any of the claimed embodiments may be used in any combination.

[0208] It should be noted that the above embodiments illustrate rather than limit the invention, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between brackets should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising several different elements and by means of appropriately programmed computers. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names.

[0209] The above is merely a description of specific embodiments of the present invention, and the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention are intended to be covered by the scope of protection of the present invention. The scope of protection of the present invention shall be based on the scope of protection of the claims.

Claims

1. An urban emergency management system, characterized in that: include: A perception module is used to collect multi-source heterogeneous data in the city in real time and pre-process the multi-source heterogeneous data to generate event situation data; An orchestration module is configured to receive the event situation data, identify events based on preset event processing rules and emergency knowledge models, and generate an emergency response process; An execution module is used to receive and analyze the emergency response process, generate specific execution instructions, and coordinate the scheduling and task allocation of required emergency resources; The execution result data of the execution module is fed back to the orchestration module for iterative optimization of the orchestration module.

2. The system according to claim 1, wherein: The perception module includes: A multimodal sensor network, deployed in urban surveillance areas, includes at least one of the following sensors: radar water level gauge, smart manhole cover, fiber optic vibration sensor, and infrared thermal imaging camera; A hybrid networking unit, including 5G and LoRaWAN heterogeneous network architecture, for data transmission of the multimodal sensor network; The data preprocessing pipeline includes a noise filtering unit for filtering out abnormal data and a spatiotemporal alignment unit for fusing data from different sources.

3. The system according to claim 1, wherein: The arrangement module includes: A composite event processing engine, built on a stream processing framework, is used to detect composite event patterns from the event situation data in real time based on predefined spatiotemporal association rules; An emergency knowledge graph, storing entities, relationships, and rules in the emergency field, for performing semantic enhancement and causal reasoning on events detected by the composite event processing engine; The process intelligent evolution engine is used to retrieve, cross and mutate from the emergency case library according to the characteristics of the incident, and generate an emergency response process that matches the current situation.

4. The system according to claim 3, characterized in that The process intelligent evolution engine uses an improved genetic algorithm to generate an emergency response process that matches the current situation. The improvements include: The weighted Jaccard coefficient is used to calculate the case similarity to screen individuals to initialize the population; Adopting a directed mutation operator guided by the emergency knowledge graph to improve the effectiveness of the mutation operation; The improved genetic algorithm uses a multi-dimensional fitness function including time cost, resource utilization and social impact to evaluate the generated emergency response process.

5. The system according to claim 3, wherein: The orchestration module also includes a process compliance verification unit, which uses linear temporal logic or TLA+ formalization method to verify the logical correctness and security before the emergency response process is executed.

6. The system according to claim 1, wherein: The execution module includes: Elastic microservice clusters encapsulate emergency response functions into independently deployable and scalable microservices; Intelligent API gateway, used to implement protocol conversion and data format adaptation between the microservice cluster and external heterogeneous business systems; The resource collaborative scheduling unit includes a multi-agent game model based on non-cooperative game theory and multiple emergency participants, which is used to solve the optimal allocation plan of emergency resources; It includes a multi-agent game model based on non-cooperative game theory, and uses a multi-objective optimization algorithm to coordinate the scheduling and task allocation of required emergency resources; Among them, the traffic scheduling of the intelligent API gateway is dynamically optimized using the Q-learning reinforcement learning algorithm; the resource collaborative scheduling unit uses the Shapley value model to quantify resource contribution and the NSGA-II algorithm to solve multi-objective optimization problems.

7. The system according to claim 1, wherein: It also includes a blockchain evidence storage module, which is built based on consortium chain technology and contains smart contracts for storing key resource scheduling records and task execution status generated by the execution module on the chain in the form of transactions, forming an unalterable emergency log; the smart contract contains a contribution quantification model, which automatically calculates the contribution score of each participating resource based on the emergency log on the chain.

8. The system according to claim 7, characterized in that Hyperledger Fabric technology is used to implement blockchain evidence storage.

9. A method for emergency management of urban emergencies, characterized in that: include: Collecting urban multi-source heterogeneous data in real time and pre-processing the multi-source heterogeneous data to generate event situation data; Receive the event situation data, perform event identification based on preset event processing rules and emergency knowledge models, and generate an emergency response process; Receive and analyze the emergency response process, generate specific execution instructions, and coordinate the scheduling and task allocation of required emergency resources; The module to be orchestrated is iteratively optimized according to the execution result data.

10. A computer-readable storage medium having computer program instructions stored thereon, wherein the computer program instructions implement the method according to claim 9 when executed by a processor.

Images